README / index.html
joeldrude's picture
Replace template with organization card
252cc2a verified
Raw History Blame Contribute Delete
2.1 kB
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1" />
<title>A2Genesis</title>
<style>
:root { color-scheme: light dark; }
body { margin: 0; background: transparent; font-family: ui-sans-serif, system-ui, -apple-system, "Segoe UI", Roboto, sans-serif; font-size: 15px; line-height: 1.6; color: #1f2937; }
h2 { font-size: 13px; text-transform: uppercase; letter-spacing: .07em; margin: 22px 0 8px; color: #111827; }
p { margin: 0 0 14px; }
ul { margin: 0 0 14px; padding-left: 20px; }
li { margin-bottom: 6px; }
a { color: #2563eb; }
code { font-family: ui-monospace, SFMono-Regular, Menlo, monospace; font-size: 13px; background: #f3f4f6; padding: 1px 4px; border-radius: 4px; }
.note { font-size: 13px; opacity: .7; }
@media (prefers-color-scheme: dark) {
body { color: #d1d5db; }
h2 { color: #f3f4f6; }
a { color: #93c5fd; }
code { background: #374151; }
}
</style>
</head>
<body>
<p><strong>A2Genesis</strong> is an independent software studio in Bochum, Germany. We build custom applications and run language models on our own hardware.</p>
<p>Here we publish quantized checkpoints of open-weight models. The quantization recipe, the calibration data and the serving configuration are documented in every model card, so a result can be reproduced rather than taken on trust.</p>
<h2>Models</h2>
<ul>
<li><a href="https://huggingface.co/a2genesis/Qwen3.8-27B-NVFP4" target="_blank">Qwen3.8-27B-NVFP4</a> &mdash; NVFP4 (W4A16) quantization of <code>Qwen/Qwen3.8-27B</code>, produced with NVIDIA TensorRT Model Optimizer 0.45. About 21 GB instead of 55 GB, so it fits on a single GPU. Apache 2.0.</li>
</ul>
<h2>Focus</h2>
<ul>
<li>Post-training quantization (NVFP4, FP8, NVIDIA ModelOpt)</li>
<li>Self-hosted inference with vLLM and SGLang</li>
<li>AI-native application development</li>
</ul>
<p><a href="https://a2genesis.de" target="_blank">a2genesis.de</a></p>
<p class="note">Not affiliated with or endorsed by NVIDIA or the Qwen team. NVIDIA and TensorRT are trademarks of NVIDIA Corporation.</p>
</body>
</html>